Prediction of Respiratory Syncytial Virus-Associated Hospitalizations Using Machine Learning Models Based on Environmental Data
Eric Guo
Head-Royce School, Oakland, United States of America
Publication date: July 10, 2026
Head-Royce School, Oakland, United States of America
Publication date: July 10, 2026
DOI: http://doi.org/10.34614/JIYRC2026I27
ABSTRACT
Respiratory syncytial virus (RSV) is a leading cause of hospitalization among young children, with outbreaks strongly influenced by environmental conditions. This study presents the first machine learning framework to integrate wastewater surveillance with meteorological and air quality data to predict RSV-associated hospitalizations in the United States three weeks ahead. Classification models were trained to predict weekly risk levels categorized as Low risk, Alert, and Epidemic. The Random Forest model achieved the best performance with a prediction accuracy of 75.8%. Wastewater RSV level was the most important predictor, followed by ozone, relative humidity, and surface pressure. Notably, significantly higher hospitalization rates were observed among American Indian and Alaska Native populations (AI/AN) and high-altitude states. These findings highlight the potential of integrating environmental and community surveillance data for early outbreak detection and more effective public health planning. An interactive R Shiny dashboard was developed to enable real-time prediction of hospitalization risk.
Respiratory syncytial virus (RSV) is a leading cause of hospitalization among young children, with outbreaks strongly influenced by environmental conditions. This study presents the first machine learning framework to integrate wastewater surveillance with meteorological and air quality data to predict RSV-associated hospitalizations in the United States three weeks ahead. Classification models were trained to predict weekly risk levels categorized as Low risk, Alert, and Epidemic. The Random Forest model achieved the best performance with a prediction accuracy of 75.8%. Wastewater RSV level was the most important predictor, followed by ozone, relative humidity, and surface pressure. Notably, significantly higher hospitalization rates were observed among American Indian and Alaska Native populations (AI/AN) and high-altitude states. These findings highlight the potential of integrating environmental and community surveillance data for early outbreak detection and more effective public health planning. An interactive R Shiny dashboard was developed to enable real-time prediction of hospitalization risk.
